lib/cp1251.h

   1 /*
   2  * Copyright (C) 1999-2001, 2016 Free Software Foundation, Inc.
   3  * This file is part of the GNU LIBICONV Library.
   4  *
   5  * The GNU LIBICONV Library is free software; you can redistribute it
   6  * and/or modify it under the terms of the GNU Lesser General Public
   7  * License as published by the Free Software Foundation; either version 2.1
   8  * of the License, or (at your option) any later version.
   9  *
  10  * The GNU LIBICONV Library is distributed in the hope that it will be
  11  * useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  * Lesser General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU Lesser General Public
  16  * License along with the GNU LIBICONV Library; see the file COPYING.LIB.
  17  * If not, see <https://www.gnu.org/licenses/>.
  18  */
  19
  20 /*
  21  * CP1251
  22  */
  23
  24 static const unsigned short cp1251_2uni[128] = {
  25   /* 0x80 */
  26   0x0402, 0x0403, 0x201a, 0x0453, 0x201e, 0x2026, 0x2020, 0x2021,
  27   0x20ac, 0x2030, 0x0409, 0x2039, 0x040a, 0x040c, 0x040b, 0x040f,
  28   /* 0x90 */
  29   0x0452, 0x2018, 0x2019, 0x201c, 0x201d, 0x2022, 0x2013, 0x2014,
  30   0xfffd, 0x2122, 0x0459, 0x203a, 0x045a, 0x045c, 0x045b, 0x045f,
  31   /* 0xa0 */
  32   0x00a0, 0x040e, 0x045e, 0x0408, 0x00a4, 0x0490, 0x00a6, 0x00a7,
  33   0x0401, 0x00a9, 0x0404, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x0407,
  34   /* 0xb0 */
  35   0x00b0, 0x00b1, 0x0406, 0x0456, 0x0491, 0x00b5, 0x00b6, 0x00b7,
  36   0x0451, 0x2116, 0x0454, 0x00bb, 0x0458, 0x0405, 0x0455, 0x0457,
  37   /* 0xc0 */
  38   0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
  39   0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
  40   /* 0xd0 */
  41   0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
  42   0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
  43   /* 0xe0 */
  44   0x0430, 0x0431, 0x0432, 0x0433, 0x0434, 0x0435, 0x0436, 0x0437,
  45   0x0438, 0x0439, 0x043a, 0x043b, 0x043c, 0x043d, 0x043e, 0x043f,
  46   /* 0xf0 */
  47   0x0440, 0x0441, 0x0442, 0x0443, 0x0444, 0x0445, 0x0446, 0x0447,
  48   0x0448, 0x0449, 0x044a, 0x044b, 0x044c, 0x044d, 0x044e, 0x044f,
  49 };
  50
  51 static int
  52 cp1251_mbtowc (conv_t conv, ucs4_t *pwc, const unsigned char *s, size_t n)
  53 {
  54   unsigned char c = *s;
  55   if (c < 0x80) {
  56     *pwc = (ucs4_t) c;
  57     return 1;
  58   }
  59   else {
  60     unsigned short wc = cp1251_2uni[c-0x80];
  61     if (wc != 0xfffd) {
  62       *pwc = (ucs4_t) wc;
  63       return 1;
  64     }
  65   }
  66   return RET_ILSEQ;
  67 }
  68
  69 static const unsigned char cp1251_page00[32] = {
  70   0xa0, 0x00, 0x00, 0x00, 0xa4, 0x00, 0xa6, 0xa7, /* 0xa0-0xa7 */
  71   0x00, 0xa9, 0x00, 0xab, 0xac, 0xad, 0xae, 0x00, /* 0xa8-0xaf */
  72   0xb0, 0xb1, 0x00, 0x00, 0x00, 0xb5, 0xb6, 0xb7, /* 0xb0-0xb7 */
  73   0x00, 0x00, 0x00, 0xbb, 0x00, 0x00, 0x00, 0x00, /* 0xb8-0xbf */
  74 };
  75 static const unsigned char cp1251_page04[152] = {
  76   0x00, 0xa8, 0x80, 0x81, 0xaa, 0xbd, 0xb2, 0xaf, /* 0x00-0x07 */
  77   0xa3, 0x8a, 0x8c, 0x8e, 0x8d, 0x00, 0xa1, 0x8f, /* 0x08-0x0f */
  78   0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, /* 0x10-0x17 */
  79   0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, /* 0x18-0x1f */
  80   0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, /* 0x20-0x27 */
  81   0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, /* 0x28-0x2f */
  82   0xe0, 0xe1, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, /* 0x30-0x37 */
  83   0xe8, 0xe9, 0xea, 0xeb, 0xec, 0xed, 0xee, 0xef, /* 0x38-0x3f */
  84   0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, /* 0x40-0x47 */
  85   0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff, /* 0x48-0x4f */
  86   0x00, 0xb8, 0x90, 0x83, 0xba, 0xbe, 0xb3, 0xbf, /* 0x50-0x57 */
  87   0xbc, 0x9a, 0x9c, 0x9e, 0x9d, 0x00, 0xa2, 0x9f, /* 0x58-0x5f */
  88   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x60-0x67 */
  89   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x68-0x6f */
  90   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x70-0x77 */
  91   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x78-0x7f */
  92   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x80-0x87 */
  93   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x88-0x8f */
  94   0xa5, 0xb4, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x90-0x97 */
  95 };
  96 static const unsigned char cp1251_page20[48] = {
  97   0x00, 0x00, 0x00, 0x96, 0x97, 0x00, 0x00, 0x00, /* 0x10-0x17 */
  98   0x91, 0x92, 0x82, 0x00, 0x93, 0x94, 0x84, 0x00, /* 0x18-0x1f */
  99   0x86, 0x87, 0x95, 0x00, 0x00, 0x00, 0x85, 0x00, /* 0x20-0x27 */
 100   0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x28-0x2f */
 101   0x89, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x30-0x37 */
 102   0x00, 0x8b, 0x9b, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x38-0x3f */
 103 };
 104
 105 static int
 106 cp1251_wctomb (conv_t conv, unsigned char *r, ucs4_t wc, size_t n)
 107 {
 108   unsigned char c = 0;
 109   if (wc < 0x0080) {
 110     *r = wc;
 111     return 1;
 112   }
 113   else if (wc >= 0x00a0 && wc < 0x00c0)
 114     c = cp1251_page00[wc-0x00a0];
 115   else if (wc >= 0x0400 && wc < 0x0498)
 116     c = cp1251_page04[wc-0x0400];
 117   else if (wc >= 0x2010 && wc < 0x2040)
 118     c = cp1251_page20[wc-0x2010];
 119   else if (wc == 0x20ac)
 120     c = 0x88;
 121   else if (wc == 0x2116)
 122     c = 0xb9;
 123   else if (wc == 0x2122)
 124     c = 0x99;
 125   if (c != 0) {
 126     *r = c;
 127     return 1;
 128   }
 129   return RET_ILUNI;
 130 }